I have a HV20 and I'll show you my wobbling... (I guess this is it) :)

ncje said:
I walked past a big camera retailer last week and decided to try another one to see I could get it to happen. SO whilst the clerks were rushing around I gave it full zoom and a shake - nothing. Wonder if the cmos chips have been altered.

Those with the wobbling issue when did you buy your camera? Be interesting to compare.
Click to expand...
Interesting, I bought mine in May. I think I even bought it the same day it was released in Sweden, which was not long after it's PAL release.
However, wouldn't it be hard to tell from watching the camera's display only?
 
Last edited:
ncje said:
I walked past a big camera retailer last week and decided to try another one to see I could get it to happen. SO whilst the clerks were rushing around I gave it full zoom and a shake - nothing. Wonder if the cmos chips have been altered.

Those with the wobbling issue when did you buy your camera? Be interesting to compare.
Click to expand...

Ncje, that's an interesting possibility. I know Canon does do this type of thing. Fail to publicly ackowledge that there is an actual problem then quietly fix it and move on. The whole XL2 fuse blowing incident seemed to be mitigated after a certain time with virtually no complaints after a while. I'm thinking, however, that the HV20 and this "wobbling" issue, are not old enough for Canon to have taken action yet.
 
I am trying to conceptualize the effect of the higher shutter speed making the situation worse and I must admit that I do not get it...

It seems to me that a faster shutter speed should make the effect less noticeable, not more. (Unless the phenomenon is just lost in and blur that motion causes at the slower shutter speeds.) Does the entire CMOS charge then dump row by row, or does the next row charge while the previous one dumps?

Like Barry stated, for professional use this would be an ideal camera to put in harm's way to get that necessary shot (which is what I bought it for). I am putting it into the worst possible situation, full of vibrations, rapid panning, and the occasional jolt. As Barry predicted I am struggling with it.

My current method involves 'floating' it in a low density, open cell foam rubber. The trick is that the camera must sink about 1/2" into the 1" foam under it's own weight. This was accomplished by cutting and enlarging a hole in the foam sheet under the mass center of the camera. I've been working on it for a few days now, lots of trial and error, but it will be worth it to have an audiance wonder "How the hell did they do that?"

I got 30 days to take it back to Best Buy. If I can make it work, great. If not it's back to the drawing board...

Bob
 
Sorry if this was posted but I didn't have time to read the whole thread but...

I did some HDV PF24 recoding from the HV20. In NTSC land the 23.98p is recorded in 59.94i with 3:2 pull-down.

I was testing Canopus ProCoder3's Inverse Telecine filter and viewed the rendered footage to Canopus HQ Codec to edit with Edius v4.24 OHCI HD 24Hz project profile.

When viewing the 23.98p non-filtered footage I noticed objects getting slightly deformed in motion like the field order blending was incorrect. But doing the same render within PC3 with the Inverse Telecine filter applied the image looked perfectly like film motion with no deformed objects.

...Angelo
 
"The IMG stabilization should not affect the image quality since it's optical"

i think it is more the optical stabilization then it being CMOS ...that based on i have a original elura and a GL1 ... when shooting in very low light hand held ( OIS turned ON ) i can get the elura image to do the wobble effect ... i've also have seen it on the GL 1 ( again very low light, OIS turned ON) .... so far i really have to shake the HV 20 to see any wobble ....
i'm not so sure that the OIS on the HV20 is totally OFF ( when it is set to off) ???
 
donatello said:
"The IMG stabilization should not affect the image quality since it's optical"

i think it is more the optical stabilization then it being CMOS ...that based on i have a original elura and a GL1 ... when shooting in very low light hand held ( OIS turned ON ) i can get the elura image to do the wobble effect ... i've also have seen it on the GL 1 ( again very low light, OIS turned ON) .... so far i really have to shake the HV 20 to see any wobble ....
i'm not so sure that the OIS on the HV20 is totally OFF ( when it is set to off) ???
Click to expand...

Now that you brought this up, I too have the Canon Elura 1999 model. My brother was going to Greece and I lent him the camcorder and some very shaky shoots I seen objects being deformed or what this thread calls wobbing. The 1999 Elura has OIS and single CCD (not CMOS). I too would say it's an OIS issue. But I doubt that I will see this problem in my footage since I keep my shots steady.

But If the OIS controller was on the fritz I can see this being a problem too.

...Angelo
 
Last edited:
The wobble issue has nothing at all to do with OIS. It has to do with the way a rolling-shutter CMOS sensor exposes. Pan horizontally past a vertical object and you'll see that object leaning when it should be vertical. That's not OIS, that's the shear introduced by a rolling shutter.
http://www.ssontech.com/content/crooked.mov

More to the point, look at the footage of the desk fan. That's a locked-down tripod shot, no OIS involved at all, yet the fan gets oddly distorted. It's because of the way CMOS works, when combined with using a rolling shutter. And it gets more exaggerated with the slower frame rate. Rolling shutter + CMOS + 24p = challenges to the footage. It's not so bad at 60i, and it doesn't happen at all with a global shutter. This footage was shot at 60i on an FX7; had it been shot at 24p the distortion would be much more.
http://www.fxsupport.de/fx.mpg

Here's an article that describes the rolling-shutter effect and how it differs from a CCD global shutter, or a CMOS chip with a global shutter (and why a rolling shutter will always be a problem when placed in high-motion circumstances).
http://www.ptgrey.com/support/kb/index.asp?a=4&q=115

Also, rolling shutter causes another problem (and one that is even referenced in the HV20's manual): if you're shooting something that involves a flash of light, the rolling shutter will expose only part of the frame. So if you're shooting something where there's flash photography going on (like maybe a red-carpet premiere or something) or nature photography of lightning, a rolling-shutter camera may fail to capture the full frame. You have to understand that the way a rolling-shutter camera exposes is basically that every line in the frame exposes at its own time, progressively from top to bottom, so the bottom line might expose as much as a full 1/24th of a second later than the top line did. So if you're shooting something that has a very brief flash of light (maybe 1/100th of a second) on a rolling-shutter camera at 24p, it's entirely possible that only 1/4th of the frame will have the flash of light, and the rest of the frame will be darker. That's entirely different from the way a global shutter camera would expose, wherein the entire frame would be lit up (capturing the full flash). If you're trying to record lightning strikes on a CMOS rolling-shutter camera you may end up being extremely disappointed.

The problem cannot have been "fixed" by Canon unless they've ditched the rolling shutter chip and replaced it with a global shutter chip. Which is possible, I guess, but I'd imagine it's highly unlikely -- this is a low-cost consumer model being sold by the tens of thousands, and the concerns of a few DVXUser filmmakers aren't likely to dictate a massive change in the production run of this little camera.
 
Last edited:
Well, looking at the HV20 poll and most of the other forums... there really doesn't seem to be any issue at all?

Maybe a small batch of bad cameras?

Myself, and many others, can't seem to create this wobble effect.

Mike
 
Kyser said:
Well, looking at the HV20 poll and most of the other forums... there really doesn't seem to be any issue at all?

Maybe a small batch of bad cameras?

Myself, and many others, can't seem to create this wobble effect.

Mike
Click to expand...


For me its not possible to notice the wobble in the cam-lcd but on computer, uhä!

What serial number do you have on your cam, is it the number behind the batteri and do you got pal or ntsc? I notice the wobble in all my clips if I not shoot stable enough.. :undecided
 
Bulan said:
For me its not possible to notice the wobble in the cam-lcd but on computer, uhä!

What serial number do you have on your cam, is it the number behind the batteri and do you got pal or ntsc? I notice the wobble in all my clips if I not shoot stable enough.. :undecided
Click to expand...


Have you connected the HV20 directly to a HDTV to see your recordings? Do you see wobbling?

...Angelo
 
angelo913 said:
Have you connected the HV20 directly to a HDTV to see your recordings? Do you see wobbling?

...Angelo
Click to expand...

No sorry dont have a hdtv but I see it when I connect it to my cumputer, guess its the same thing. Going to upload a clip shortly of my wobble.. :embarasse
 
Last edited:
My HV20 Warbled! Wobbled?

My HV20 Warbled! Wobbled?

Yay, I had a warble! Or is it: booo, my camera warbles too? I haven't read all 12 pages of posts yet, but I do know my HV20 did a lovely warble yesterday. It was in Program mode shooting SD DV, and it did the warble while panning past a bright area (outdoors). It was visible while reviewing the footage in camera, and I just saw it again while capturing the video to computer.
 
Last edited:
seunosewa said:
if you have to use abnormal settings to see the problem, it's not a real problem
Click to expand...


Just bought an hv30 two days ago to use as a helmet cam and a capture deck.

At 24p, cine mode, and 1/60 shutter, my footage from the helmet looks like I got the bonus jello lens adapter as well.

Not the camera for my purposes it looks like.



so....................any other compact/consumer cameras out there that shoot 24p without a rolling shutter?
 
You need a CCD camera, you have to avoid CMOS. Look at the Panasonic SD9, it's a micro-sized AVC-HD camcorder that uses CCD. Won't work as a "capture deck" though...
 
Thanks for the response Barry. The SD9 is certainly on the short list. From what I can tell this camera is presenting some problems with fast panning motions. For use as a helmet mounted POV camera this might prove to be utterly useless since every shot will essentially be one continuous panning motion. At the moment it seems I'm stuck between warping and stuttering distortions if I want 24p in a small package. I'm wondering if the stuttering is a pulldown issue that has a workaround.

The size of the sd9 makes me outright giggle though.
 
There are also situations where EIS is better than OIS.

OIS delivers a cleaner and more detailed picture because the camera records exactly what hits the photo sites on the sensor whereas EIS may skew the image and interpolate pixels.

However, in extreme situations it is possible to overwhelm the mechanical OIS system. Strong vibrations and oscillations are good examples -- if you mount a camera to a motorcycle, an offroad vehicle or an aircraft, you may get better results with EIS.
 
THoff said:
There are also situations where EIS is better than OIS.

OIS delivers a cleaner and more detailed picture because the camera records exactly what hits the photo sites on the sensor whereas EIS may skew the image and interpolate pixels.

However, in extreme situations it is possible to overwhelm the mechanical OIS system. Strong vibrations and oscillations are good examples -- if you mount a camera to a motorcycle, an offroad vehicle or an aircraft, you may get better results with EIS.
Click to expand...


You know any of the smaller hd consumer camcorders that are worth damn that use an EIS?

Seems like once the gospel was stated that OIS is superior, all consumer cams went that way on the "upper end" versions.
 
You must log in or register to reply here.
Back
Top